Multiple payload slicer system with prenormalization integer...

Details Multiple payload slicer system with prenormalization integer... Multiple payload slicer system with prenormalization integer...

2000-03-28

2003-12-09

Phu, Phoung (Department: 2631)

Pulse or digital communications

Systems using alternating or pulsating current

Plural channels for transmission of a single pulse train

C375S320000

Reexamination Certificate

active

06661849

ABSTRACT:

TECHNICAL FIELD
The present invention relates generally to network interfacing, and more particularly, to an apparatus and method for slicing a received modulated signal including quadrature amplitude modulated signal to recover transmitted data.
BACKGROUND OF THE INVENTION
The transmission of various types of digital data between computers continues to grow in importance. A predominant method of transmitting such digital data includes coding the digital data and modulating a high frequency carrier signal in accordance with the coded digital data. A coding and modulation technique known as quadrature amplitude modulation provides for modulating both the carrier amplitude and the carrier phase to represent encoded data. The QAM modulated high frequency carrier signal is transmitted across a network physical transmission medium such as electrical cable, fiber optic, RF, or other medium to a remote computing station.
At the remote computing station, the high frequency carrier signal must be received, demodulated, and decoded, including slicing, to recover the original data. In the absence of any distortion across the network medium, the received signal would be identical in phase, amplitude, and frequency to the transmitted carrier and could be demodulated and decoded using known techniques to recover the original data.
One problem with networks is that the physical medium and network topology tend to distort the carrier signal, especially at high frequencies. Branch connections and different lengths of such branches cause reflections of the transmitted signal. Such problems are even more apparent in a network which uses home telephone wiring cables as the network physical medium. The typical wiring of the telephone network is designed for the “plain old telephone service” (POTS) signals in the 3-10 kilohertz frequency and are not designed for transmission of high frequency carrier signals in a frequency range greater than 1 MHz. The high frequency carrier signal is also distorted by transients in wiring characteristics due to on-hook and off-hook switching and noise pulses of the POTS (e.g. ringing). The high frequency carrier is further distorted by spurious noise caused by electrical devices operating in close proximity to the “cable” medium.
Such distortion of frequency, amplitude, and phase of the high frequency carrier signal degrades network performance and tends to impede the design of higher data rate networks.
One technique for compensating for such distortion would be to slow the data rate by using a lower payload encoding. For example, in an 8 data bits/baud payload encoding system, the receiver must distinguish between 256 distinct constellation coordinates, each representing a distinct combination of a carrier signal phase and amplitude. Each constellation coordinate corresponds to 8 bits of data information and is typically called a symbol. Alternatively, in a 4 data bits/baud payload encoding system, the receiver only needs to distinguish between 16 distinct constellation coordinates and in a 2 bits/baud payload encoding system, the receiver only needs to distinguish between 4 distinct constellation coordinates. The 2 bits/baud system will be more tolerant to distortion because the distorted carrier phase and amplitude are less likely to mis-map to an incorrect constellation coordinate.
More specifically, referring to
FIGS. 1
a
and
1
b
, a known 2 bits/baud constellation
20
(2 bits per symbol) and a 4 bits/baud constellation
22
(4 bits per symbol) are shown respectively. Constellation
20
includes 4 defined constellation coordinates
24
(
a
)-
24
(
d
). Each constellation coordinate
24
(
a
)-
24
(
d
) represents a data symbol—a unique combination of carrier magnitude and phase. For example, the magnitude of vector
26
and the phase angle
28
correspond to constellation coordinate
24
(
b
).
Constellation
22
includes 16 defined constellation coordinates. Again, each coordinate represents a unique combination of carrier magnitude and phase. For example, the magnitude of vector
32
and the phase angle
34
correspond to constellation coordinate
30
(
a
) and the magnitude vector
36
and the phase angle
38
correspond to constellation coordinate
30
(
b
).
It should be appreciated that both constellation
20
and
22
are QAM Square constellations in that a perimeter of an area bounded by the coordinates is square. Constellation
22
, representing 2 bits/baud is actually a phase shift keying (PSK) modulation because the amplitude of all coordinates is the same. Similarly, a 3 bits/baud constellation (constellation) is also PSK.
In operation, a transmitter may transmit a modulated carrier with a particular phase and magnitude corresponding to 2/bits baud coordinate
24
(
a
). Due to distortion, the receiver may detect a carrier phase and magnitude corresponding to point
40
. The receiver must determine to which of the constellation coordinates
24
(
a
)-
24
(
d
) the received point
40
corresponds in order to recover the 2 bits of transmitted data. Using a 2 bits/baud transmission, any received point within shaded area
42
will map to constellation coordinate
24
(
a
). Furthermore, any received points outside the shaded area in the same quadrate will be rounded to map to constellation coordinate
24
(
a
).
However, in a 4 bits/baud transmission, a transmitter may transmit a carrier with a phase and magnitude corresponding to point
30
(
c
) which happens to have the same magnitude and phase as the 2 bits/baud coordinate
24
(
a
). And, because of the same distortion, the receiver detects a carrier phase and magnitude corresponding to point
40
. While this distortion was tolerable in a 2 bits/baud transmission, the distortion causes a mis-map in a 4 bits/baud transmission because the received point
40
is within shaded area
44
which will map to coordinate
30
(
d
)—not the originally transmitted coordinate
30
(
c
).
While it is therefore obvious that the lower payload encoding would be more tolerable to distortion, slowing the data rate of all transmissions on a network to overcome the worst distortion has the drawback of reducing network throughput.
Therefore, a recognized solution is to use adaptive payload encoding wherein, based on carrier distortion between a transmitter and a receiver, the maximum payload encoding can be selected which still enables the receiver to properly distinguish between combinations of carrier phase and amplitude. Therefore, if there is relatively little carrier distortion between a particular transmitter and a particular receiver, the two will negotiate a large payload encoding, such at 8 bits per baud for rapid data transmission. However, if there is significant carrier distortion between a particular transmitter and a particular receiver, the two will negotiate a smaller payload encoding, such as 2 bits per baud, to assure error free transmission.
It is desirable that the maximum carrier amplitude is the same regardless of which payload encoding is used, typically, the magnitude of both the I-value and the Q-value of the four outermost constellation coordinates in each of the QAM Square constellations (e.g. 2 bits/baud (PSK), 4 bits/baud, 6 bits/baud, and 8 bits/baud) a value equal to the square root of one half of the maximum amplitude squared and each inner coordinate to a fractional value less than such values. However, an adequate number of bits (for example 18 bits) of precision would be needed to map a symbol in the constellation tables to maintain an adequate signal to noise ratio for reliably recovering the symbol in receiver. Merely reducing the precision by one bit would decrease the signal to noise ratio by 6 db.
The problem is that in such an adaptive environment where payload encoding options include 2, 3, 4, 5, 6, 7, or 8 bits per baud, both the transmitter and the receiver must be able to accommodate all payload options. Such adaptive systems are typically implemented by digital signal processing (DSP) which enables the calculations to be performed quickly enough to transmit and receive the data.

Affiliated with

Also associated with

Rating

Multiple payload slicer system with prenormalization integer... does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multiple payload slicer system with prenormalization integer..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multiple payload slicer system with prenormalization integer... will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3177925